1870
COMMUN~CAT~ONS TO THE EDITOR
Vol. 74
converted to cholesterol by way of 2-carbon intermediates.
ultraviolet absorption maxima of I gives evidence that the two recognized acetylenic bonds are in conjugation.8 The remaining two units of unsaturaDEPARTMENT OF BIOCHEMISTRY UNIVERSITY OF CHICAGO R. G.LANGD0Nl3 tion are believed to be a conjugated diene, conCHICAGO, ILLINOIS KONRAD BLOCH jugated with the allenic group but not with the 9, 1952 RECEIVED FEBRUARY carboxyl group. The location of the observed long wave length (13) U . S. Public Health Service Research Fellow. ultraviolet absorption maximum of I at 281 r n N limits the length of contributing conjugated multiple bonds to three units, part of which may be THE STRUCTURE OF THE ANTIBIOTIC MYCOMYCIN a~etylenic.~The allene bond in the proposed 7,8position of I performs a unique chromophoric role Sir : We have deduced the structure of the highly whereby the A? bond and the As bond are conjuunstable antibiotic mycomycin' as 3,5,7,8-tridecatet- gated with the 3,5-diene and 10,12-diyne, respectively, while the central carbon atom of the allene raene-lO,l2-diynoic acid (I). serves as an electronic insulator between the two HCEC-C%CCH=C=CHCH=CH-CH-CHCHCH~CO~H resulting conjugated systems.5 The two effecMycomycin undergoes an unusual rearrangement tively isolated chromophores, each totaling three in normal aqueous potassium hydroxide at 27' units of conjugation, explain the observed general involving an allene to acetylene isomerization ac- light absorption zone of I. The infrared spectrum of I1 in dioxane exhibits companied by migration of existing acetylenic bonds. The rearranged acid, isomycomycin, has characteristic bands a t 2200 ern.-' and 1730 cm.-l and unconbeen assigned the structure 3,5-tridecadiene-7,9,11- attributed to disubstituted -C=Cjugated -COzH,2a respectively. Monosubstituted triynoic acid (11). acetylenic and allenic bands are absent. I1 analyzes CH&=C-CIC-CX-CH=CH-CH=CH-CH2-COZH for one C-methyl group, whereas I possesses none. I crystallizes in colorless needles from methylene I1 does not react with alcoholic silver nitrate, funcchloride at -40'; m.p. 75' (dec. explosively) and substantiating the absence of a =C-H [ c Y ] ~ ~-130' D (c, 0.4, ethanol). (Anal. Calcd. for tion. The ultraviolet light absorption properties CI~HIOO~: C, 78.76; H, 5.08; one C-methyl, 7.6; of I1 prove to be very similar to those recorded neut. equiv., 198; Found: C, 78.17; H, 5.36; for compounds containing a conjugated dieneI1 reacts with ethereal diazoC-methyl (Kuhn-Roth), 0.48; neut. equiv. 200). triyne g r o ~ p i n g . ~ ~ ~ methane to form a methyl ester, crystallized from Ultraviolet absorption in diethyl ether: A:$, 226, ether-hexane as colorless needles, m.p. 69-70'. A e 35,000; >:Ex. 267, E 61,000; A&!:. 281, E 67,000. Diels-Alder reaction of the methyl ester of I1 with I1 crystallizes in colorless needles from ether- maleic anhydride gives a monoaddition product hexane, decomposes slowly above 140' and is (III), crystallized from acetone-hexane as colorless optically inactive. (Anal. Found: C, 78.87; H, plates, m.p. 177-178' (dec.). (Anal. Calcd. for 5.43; C-methyl (Kuhn-Roth), 9.6; neut. equiv., Cl8HI4O6:C, 69.67; H, 4.55; CHsO, 10.00. Found: 198). Ultraviolet absorption in diethyl ether : C, 69.59; H, 4.64; CH,O, 10.89.) Ultraviolet light &:A 246, E 24,000; 257.5, B 58,000; LEEx. absorption in methanol: LE$. 215, E 82,000; 267, e 110,000; Xg:= 287.5, E 14,000; A,":., 305.5, 272.5, e 450; A=&, 289, e 430; 310, e 27,000; >E!&. 324, e 41,000; 347, E 34,000. e 170. These light absorption properties of I11 Complete hydrogenation of I and I1 requires /CH=CH eight moles of hydrogen, quantitatively yielding / \CH-CH2-COL!H, n-tridecanoic acid, unequivocally identified by com- CH,-C=C-C%C-CZS-CH parisons with an authentic sample.2 \CH-CH I I I11 The infrared spectrum of I in dioxane has characo=c c=o teristic bands near 3180,2200,1930 and 1730 em.-' attributed to =C-H, disubstituted -C=C--, \o/ . -CH&CHand unconjugated - - C O z H 2 a functions, respectively. I reacts with acetylenic bear striking resemblance to the unique light abhydrogen reagents such as alcoholic silver ni- sorption behavior of the conjugated triacetylene trate. In view of the linear nature of the re- grouping when it is the sole contributing chromoduction product, the high order of optical activity (3) T. Brunn, C. J. Haug and N. A. Sorensen. Acta Chem. Scand., of I can only be reconciled with its allefiic function. 4, 850 (1950). (4) I. M. Heilbron, E. R. H. Jones and R. A. Raphael, -7. Chcm. The fine structure spacing (Av', 1900 cm.-l) of the ( 1 ) E. A. Johnsonand K.L. Burdon, J . Bacr., 64.281 (1947) (2) Obtained through the courtesy of Dr H. J. Harwood or Armout and Company. (2a) In a cornpariaon of infrared spectra determined in dioxane solution, the location of the C-0 stretching frequency exhibited by I and I1 near 1730 c m - 1 is the same as the corresponding band of their respective perhydro derivatives and authentic %-ttidecairbc acid. If I and/or I1 contdned a conjuaated carboxyl group, this band would be expected t o occur a t a measurably lower frequencg than the corresponding saturated derivative.
SOC.,268 (1943); I. M. Heilbron, E. R. H. Jones and F. Sondheimer, ibid.. 1586 (1947). (5) In a somewhat analogous situation, the ultraviolet light absorption of tetraphenylallene (Xzfx. 267, 12,000) corresponds t o that of 1,l-diphenylethylene (gfxx. 250, e 11,000) and not to Iranr,lran~-diphenyl-1.3-butadiene 328, e 66,000). (6) K. Stavholt and N. A. Sorensen, Acre Chcm. Scand.. 4, 1567 (1950). (7) E. R. H. Jones, M. C. Whiting, J. 8. Armitage, C. L. Cook and
N.Bntwide. NOrwr, 168,800(1961).
April 5, 1952
COMMUNICATIONS TO THE EDITOR
1871
resisted all attempts at crystallization. A mixture of the p-phenylphenacyl esters of the acids after chromatography on a silicic acid4 column, yielded p-phenylphenacyl acetate, m.p. 110.8-1 11.2' (calcd. C, 75.58; H, 5.55; found, C, 75.38; H, 5.66) (9) E. R. H. Jones, personal communication. and p-phenylphenacyl a-methylbutyrate, m.p. 70RESEARCH LABORATORIES WALTERD. CELMER 71' cor. (calcd. C, 77.01; H, 6.80; Found: C, CHAS.PPIZERAND Co., INC. I. A. SOLOMONS BROOKLYN 6, NEWYORK 76.61; H, 6.80). RECEIVED FEBRUARY 28, 1952 The hypotensive activity6 of neogerrnitrine and escholerine in anesthetized dogs was found to be 0.13 pg. [0.12-0.15] and 0.30 pg. [0.26-0.361, HYPOTENSIVE ALKALOIDS OF VERATRUM respectively. ESCHSCHOLTZII The isolation procedure has, in addition, yielded Sirs: the alkaloids isorubijervine, jervine, rubijervine, In view of the present interest in the Veratrum pseudojervine and veratramine, already known Alkaloids as hypotensive agents, we wish to report as constituents of Veratrum eFiride, and small the isolation from Veratrum eschscholtzii Gray amounts of four apparently new crystalline alkof neogermitrine,' and a new alkaloid, for which we amines and a new ester alkaloid which will be propose the name escholerine. described more fully in a subsequent publication. Preliminary fractionation of a crude chloroform (4) J. G. Kurchner, Arthur H. Prater and A. J. Haagen-Smit, I n d . extract which was based on the selective solubilities Eng. Chem., A n a l . E d . , 18 31 (1946). of the alkaloids and their salts, in conjunction with (5) Expressed as micrograms per kilogram of dog per minute reassays2 for hypotensive activity in anesthetized quired lor a ten minute intravenous infusion to lower the mean arterial dogs, yielded an amorphous fraction that accounted blood pressure 30% when administered according to the method of G. L. Maison and J. W. Stutzman The bracketed numbers express for the major part of the hypotensive activity in the the 95% confidence limits. crude extract. Further fractionation by two 8LABORATORIES, INC. M. W. KLOHS plate Craig countercurrent distributions3 yielded RIKER 8480 BEVERLY BLM. F. KELLER two fractions, A and B, each with a high hypo- Los ANGELES,CALIFORNIA S. KOSTER tensive activity. Fraction A was resolved on a 24W. MALESH RECEIVED JUNE18, 1951 plate distribution using 2 M acetate buffer at PH 5.5 and benzene as the solvent system. Neogerrnitrine was obtained from the material recovered from tubes 8 to 13 by crystallizing from acetone-water MICROBIOLOGICAL OXYGENATION OF STEROIDS AT CARBON 1 1 (m.p. 234-234.8 (cor.)); -79 2', ( c 0.9 in pyridine); the sample was further identified by Sir: comparison of its infrared spectrum, and by a It is generally acknowledged that the most diffimixed melting point with an authentic sample of cult series of steps in the partial synthesis of cortineogermitrine from Veratrum viride Aiton kindly sone is that concerned with the introduction of provided by Dr. J. Fried. oxygen at carbon 11 of the steroid n u c l e u ~ . ~ * ~ * ~ ~ ~ Fraction B was distributed on a 24-plate Craig We wish to report the oxygenation of steroids, e.g., apparatus, using 0.5 M acetate buffer p H 5.0 and progesterone, a t carbon 11in a single step by means benzene-cyclohexane 25 :75 as the immiscible of common molds of the order Mucorales after a phases. The material recovered from tubes 8 to transformation period of 24-48 hours, in a lactal13, when crystallized from acetone-water, yielded bumin digest-dextrose-cornsteep medium. Thus, escholerine (m.p. 235-235.3 with dec. (cor.) ; from progesterone, a new 11-oxygenated steroid [ a ] 2 5 ~ -30 i 2' (c 1.0 in py.); $7 f 2' (c 1.0 intermediate is made available for conversion to the in CHC13)). The analytical data indicate the cortical hormones. In these studies we have made empirical formula C41H~10~3N; (calcd. C, 63.46; use of the procedure of Zaffaroni, et aLj5for characH, 7.92; N, 1.80; eq. wt., 775.9; found: C, terization of the transformation products. 63.42, 63.50; H, 8.00, 7.97; N, 2.04; eq. wt., 782, The ability of several micro-organisms to oxidize 772; picrate, m.p. 259.5' (dec.), (C41H61013N. a hydroxyl group or reduce a ketone group in a HOC6Ha(N0&: C, 56.17; H, 6.42; found: C, steroid is well recognized,6 but heretofore the only 56.41; H, 6.38); aurichloride, m.p. 191.4' (froth- microbiological oxygenation of a steroid carbon ing), (C41H61013N.HAuC14. C, 44.13; H, 5.GO; atom was reported by K r h l i and Horv&th7in the Au, 17.67; found: C, 44.53; H, 5.61; Au, 17.21). B. F. McKenzie, V. R. Mattox, L. L. Engel and E. C.Kendall, Volatile acid determination, found : 3.7 equivalents J . (1) Biol. Chcm., 173, 271 (1948). of acid. (2) E. M. Chamberlain, W. V. Ruyle, A. E. Erickson, J. M. ChemHydrolysis of escholerine with 0.1 N methanolic erda, L. M Aliminosa, R. L. Erickson, G.E. Sita and Max Tishler, potassium hydroxide afforded acetic acid, a- THISJOURNAL,73, 2396 (1951). (3) L. F . Fieser, J. E. Herz and Wei-Yuan Huang, ibid., 73, 2397 methylbutyric acid and a base that has so far (1951).
phore.819 More detailed accounts of the above structural considerations will appear in subsequent publications. (8) F . Bohlmann, Chcm. Bcr., 84, 786 (1951).
1
*
( 1 ) J. Fried and P . Numerof, Abst. 119th meeting A.C.S., Clxeland, Ohio April, 1951, 12L. (2) Assays were run according to the method of 0. L. Maison and J . W . Stutzman, Arch. Inf. Pharmacodyn., 85, 357 (1961). Evaluations were made a t Boston University School of Medicine, Boston, Mass. (3) J . Fried, H. White and 0.Wintersteiner, THISJOURNAL, 71, 46214630 (1950).
(4) G. Stork, J. Romo, G. Rosankranz and C. Djerassi, ibid., 73, 3546 (1951). (5) A. Zaffaroni, R. B. Burton and E. H . Keutmann, Science, ill, 6 (1950). (6) M. Welsch and C. Heusghem, Comfit. rend. sot. biol.. 148, 10741076 (1948) (7) A. Krbmli and J. Horvdth, Nature. 160, 639 (1947); 163, 219 (1949).
